Boundary-Aware Uncertainty for Feature Attribution Explainers
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2022
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913252395450368 |
|---|---|
| author | Hill, Davin Masoomi, Aria Torop, Max Ghimire, Sandesh Dy, Jennifer |
| author_facet | Hill, Davin Masoomi, Aria Torop, Max Ghimire, Sandesh Dy, Jennifer |
| contents | Post-hoc explanation methods have become a critical tool for understanding black-box classifiers in high-stakes applications. However, high-performing classifiers are often highly nonlinear and can exhibit complex behavior around the decision boundary, leading to brittle or misleading local explanations. Therefore there is an impending need to quantify the uncertainty of such explanation methods in order to understand when explanations are trustworthy. In this work we propose the Gaussian Process Explanation UnCertainty (GPEC) framework, which generates a unified uncertainty estimate combining decision boundary-aware uncertainty with explanation function approximation uncertainty. We introduce a novel geodesic-based kernel, which captures the complexity of the target black-box decision boundary. We show theoretically that the proposed kernel similarity increases with decision boundary complexity. The proposed framework is highly flexible; it can be used with any black-box classifier and feature attribution method. Empirical results on multiple tabular and image datasets show that the GPEC uncertainty estimate improves understanding of explanations as compared to existing methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2210_02419 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Boundary-Aware Uncertainty for Feature Attribution Explainers Hill, Davin Masoomi, Aria Torop, Max Ghimire, Sandesh Dy, Jennifer Machine Learning Post-hoc explanation methods have become a critical tool for understanding black-box classifiers in high-stakes applications. However, high-performing classifiers are often highly nonlinear and can exhibit complex behavior around the decision boundary, leading to brittle or misleading local explanations. Therefore there is an impending need to quantify the uncertainty of such explanation methods in order to understand when explanations are trustworthy. In this work we propose the Gaussian Process Explanation UnCertainty (GPEC) framework, which generates a unified uncertainty estimate combining decision boundary-aware uncertainty with explanation function approximation uncertainty. We introduce a novel geodesic-based kernel, which captures the complexity of the target black-box decision boundary. We show theoretically that the proposed kernel similarity increases with decision boundary complexity. The proposed framework is highly flexible; it can be used with any black-box classifier and feature attribution method. Empirical results on multiple tabular and image datasets show that the GPEC uncertainty estimate improves understanding of explanations as compared to existing methods. |
| title | Boundary-Aware Uncertainty for Feature Attribution Explainers |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2210.02419 |